Artificial intelligence can help doctors diagnose life-threatening diseases: Here are five examples

Artificial intelligence (AI) and machine learning (ML) may soon overtake doctors to correctly diagnose conditions from breast cancer to kidney disease.

Myupchar January 13, 2020 14:04:42 IST
content powered by
Artificial intelligence can help doctors diagnose life-threatening diseases: Here are five examples

Artificial intelligence (AI) and machine learning (ML) may soon overtake doctors to correctly diagnose conditions from breast cancer to kidney disease. 

Case in point: recently, a team of researchers at the University of California Los Angeles (UCLA) developed an AI model that was better at differentiating between “preinvasive lesions of the breast” — abnormal growths in breast cells — than doctors with years of experience. They published their findings in the JAMA Network Open, a science journal, on 9 August.

Artificial intelligence can help doctors diagnose lifethreatening diseases Here are five examples

Representational image. Reuters

Machine learning is the capacity of a machine to learn from its mistakes, just like people do. Machines are also able to compute and assess different scenarios much faster, and more accurately than people. 

Incorrect breast cancer diagnoses are still quite common, leading to anxiety and depression in women.

Researchers at UCLA used AI to tell the difference between two very similar looking abnormalities in the breasts: ductal carcinoma in situ (abnormal cellular growth that is restricted to the milk ducts) and atypia (in which breast cells change in size, number or shape, and which may or may not lead to cancer). Diagnosing these conditions involves a great amount of sophisticated data, interpreting which is both difficult and tricky. 

“These results are very encouraging,” said Dr Joann Elmore, the corresponding author of the UCLA research. “There is low accuracy among practising pathologists in the U.S. when it comes to the diagnosis of atypia and ductal carcinoma in situ, and the computer-based automated approach shows great promise,” she added.

Inspired by this success story, here’s a quick look at four other conditions where AI diagnosis is making significant strides:

Lung cancer

According to a study published in Nature Medicine in May 2019, Google AI researchers have used end-to-end deep learning models to predict lung cancer in patients. Using a patient’s lung cancer malignancy-risk score, they were able to identify nodules and classify them as benign or malignant.

Mortality in heart disease

In 2018, researchers at the London-based Francis Crick Institute used AI, as well as data from over 80,000 patients and variables like age, smoking habit and home visits by doctors to predict the chances of death in people living with heart disease. The AI model gave more accurate results by accessed more data points than doctors across the U.K. For example, the machine model used 600 variables, whereas the doctors used 27 to predict chances of death. 

Kidney disease

Scientists at Google-owned company DeepMind have developed an ML model that is said to accurately predict acute kidney injury 48 hours ahead of the current best diagnostic methods. Acute kidney injury is a life-threatening disorder in which a patient's kidneys stop working suddenly. The prediction can help doctors to take appropriate steps and potentially save lives.

Alzheimer’s disease

In 2018, scientists at the University of California San Francisco (UC San Francisco) developed an AI system that could detect Alzheimer’s six years before the onset of symptoms. “One of the difficulties with Alzheimer’s disease is that by the time all the clinical symptoms manifest and we can make a definitive diagnosis, too many neurons have died, making it essentially irreversible,” lead researcher Dr Jae Ho Sohn, told the UC San Francisco university newsletter.

Early detection could help medical practitioners arrest the development of the disease. Globally almost 50 million people are living with dementia - this number is expected to reach 82 million by 2030 and 152 million by 2050, according to the World Health Organization. 

(Full disclosure: myUpchar is developing AI models to detect Malaria from blood samples, and also to read chest X-rays to detect up to 13 respiratory and other diseases.)

Health articles in Firstpost are written by myUpchar.com, India’s first and biggest resource for verified medical information. At myUpchar, researchers and journalists work with doctors to bring you information on all things health. To know more about breast cancer, please visit https://www.myupchar.com/en/disease/breast-cancer

Updated Date:

also read

CES 2021: Samsung 110-inch micro LED, JetBot 90 AI+ robot, Galaxy Upcycling at Home, more announced
News & Analysis

CES 2021: Samsung 110-inch micro LED, JetBot 90 AI+ robot, Galaxy Upcycling at Home, more announced

At the event, Samsung showcased a number of products from AI robots, AIoT home devices, to solutions for a sustainable future.

Love in the time of algorithms: Would you let artificial intelligence choose your partner?
News & Analysis

Love in the time of algorithms: Would you let artificial intelligence choose your partner?

With AI-systems as matchmakers, in the coming decades it may become common to date a personalised avatar.

Turn back time? Study hints at cause, possible solution to halt decline due to ageing, Alzheimer's
Health

Turn back time? Study hints at cause, possible solution to halt decline due to ageing, Alzheimer's

Research published in the journal Nature pinpoints the role of a messenger hormone called prostaglandin E2 (PGE2), found in much higher levels in older people and mice than their younger counterparts